D2.8 Development of UCO specificationsvegetable oil is suitable for commercial process of hydrotreating with no precautions than other traditional vegetable oils. Used cooking oil

ITAKA Deliverable D2.8 / Date 30/04//2016 / Version: 0.1

Page 1 of (40) No part of this report may be used, reproduced and/or disclosed in any form or by any means without the prior written permission of the ITAKA project partners. © 2016 – All rights reserved

ITAKA Collaborative Project

FP7 – 308807

D2.8 Development of UCO

specifications Main author: Marco Buffi (RE-CORD)

With contributions from: David Chiaramonti, Giulia Lotti (RE-CORD)

Project title: Initiative Towards sustainable Kerosene for Aviation

Deliverable nature: R

Dissemination level: PU

Start date of the project 1st November 2012

Duration 48 months

Contractual delivery date: 30th April 2016

Actual delivery date: 30th April 2016

Status: Not Submitted

Contractual: Yes

Version: 0.1

Total number of pages: 33

WP number: WP2

Leader of WP: NESTE RENEWABLE FUELS OY (NRF)

Lead Beneficiary of deliverable: RECORD

Keywords: Used cooking oil, biojet fuel, drop-in, catalytic conversion, nitrogen.

This project has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement No 308807.



Executive summary

The objective of the ITAKA project is to support the development of aviation biofuels in an economically,

socially and environmental sustainable manner, improving the existing technology and infrastructures. This

report investigates the use of used vegetable oil (UCO) as renewable feedstock for the production of drop-in

renewable aviation fuels. The development of UCO specifications is crucial to allow the use a lipid waste for

commercial hydrotreating process. Raw UCO needs to be filtered and upgraded before its use as feedstock,

so the esterification was considered as upgrading treatment: three samples of esterified UCO was analysed

and investigated as potential feedstock. High grade filtration reduced contaminants and water content of raw

UCO, restoring a large number of properties to initial values of vegetable oils. Then RE-CORD investigated

the catalytic pyrolysis of UCO as alternative thermochemical treatment to commercial hydrotreating to

produce bio-hydrocarbons suitable for renewable aviation fuels. Best results (35 wt% of bio-hydrocarbons

contained into collected bio-oil) was achieved using activated carbons as catalysts, at 500 °C and WHSV of

2.5 h-1

. The experience with experimental activities on raw UCO, its upgrading and thermochemical

processing, permitted to develop a proposal for UCO specifications.



Table of contents

Executive summary ........................................................................................................................................... 2

Document Information ....................................................................................................................................... 3

Table of contents ............................................................................................................................................... 4

List of figures and tables .................................................................................................................................... 5

Abbreviations ..................................................................................................................................................... 6

Definitions .......................................................................................................................................................... 7

Introduction ................................................................................................................................................. 8 1

UCO as feedstock for biofuels production ................................................................................................ 10 2

2.1 UCO vs traditional vegetable oils ...................................................................................................... 11

2.2 Industrial esterification of UCO ......................................................................................................... 11

Lipids to biofuels pathways ...................................................................................................................... 13 3

3.1 Hydroprocessing of vegetable oils .................................................................................................... 14

3.2 Experimental pathways ..................................................................................................................... 15

3.2.1 Catalytic pyrolysis of lipids ......................................................................................................... 15

Experimental campaign on catalytic pyrolysis of UCO ............................................................................. 17 4

4.1 Goal and Scope................................................................................................................................. 17

4.2 Materials and methods ...................................................................................................................... 17

4.2.1 Pyrolysis unit .............................................................................................................................. 17

4.2.2 Feedstock ................................................................................................................................... 18

4.2.3 Methods of analysis ................................................................................................................... 19

4.2.4 Catalysts .................................................................................................................................... 21

4.2.5 Experimental runs ...................................................................................................................... 21

4.3 Results .............................................................................................................................................. 22

4.3.1 Oil-based feedstock ................................................................................................................... 22

4.3.2 Results from first experimental campaign .................................................................................. 23

4.3.3 Results from second experimental campaign ............................................................................ 26

4.3.4 Investigation of unidentified fraction of bio-oil ............................................................................ 29

4.3.5 Discussion .................................................................................................................................. 31

Development of UCO specifications......................................................................................................... 32 5

5.1 Hydrotreating barriers and constraints .............................................................................................. 32

5.2 A proposal for UCO specifications .................................................................................................... 33

Conclusions .............................................................................................................................................. 36 6

References ............................................................................................................................................... 37 7



List of figures and tables

Figure 1: RE-CORD experimental activities on UCO. ....................................................................................... 9

Figure 2: On the left side, raw UCO. On the right side, the residual of centrifugation (left) and the separated

oil (right). ................................................................................................................................................... 10

Figure 3: Thermochemical conversion routes from lipids to biofuels [29]. ...................................................... 14

Figure 4: Scheme of the main reactions in the hydrotreatment of generic vegetable oils and fatty acids [31].

.................................................................................................................................................................. 14

Figure 5: Neste Oil’s renewable diesel plant in Singapore, with a capacity of 800,000 tons per year. Source:

neste.com ................................................................................................................................................. 15

Figure 6: Reaction scheme for the pyrolysis of saturated and unsaturated triglycerides proposed by Idem et

al [39]. ....................................................................................................................................................... 16

Figure 7: A scheme for RE-CORD catalytic pyrolysis of triglycerides-based materials. ................................. 17

Figure 8: A representative scheme of the catalytic pyrolysis unit for UCO conversion. .................................. 18

Figure 9: (a) back view of the reactor and condensation system during experimental campaign; (b) bioliquid

feeding system during operation. ............................................................................................................. 26

Figure 10: GC-FID spectrum from UCO AC* test. ........................................................................................... 29

Figure 11: GC-simulated distillation. ................................................................................................................ 31

Table 1: Esterified UCO analysis (methods of analysis are described in 4.2.3). ............................................ 12

Table 2: Evaluation of properties of the adopted catalysts.............................................................................. 21

Table 3: Description of experimental activities. ............................................................................................... 22

Table 4: Analytical analysis on feedstock . ...................................................................................................... 23

Table 5: Fatty acids profile of adopted feedstock. ........................................................................................... 23

Table 6: First run yields of catalytic campaign at 500 °C. ............................................................................... 24

Table 7: Analysis on bio-oil quality for each catalytic test. .............................................................................. 25

Table 8: Composition of bio-oil fractions on first run tests analyzed by GC MS/FID. ..................................... 25

Table 9: Second run yields of catalytic campaign. UCO AC test consists in precedent test reported in Table

6. Asterisk (*) means tests carried out with increased catalysts mass. ................................................... 27

Table 10: Analysis on bio-oil quality for catalytic tests at reduced WHSV. UCO AC test consists in precedent

test reported in Table 6. Star (*) means tests carried out with increased catalysts mass. ...................... 27

Table 11: Composition of bio-oil fractions on second run tests analyzed by GC MS/FID. UCO AC test

consists in precedent test reported in Table 6. Asterisk (*) means tests carried out with increased

catalysts mass. ......................................................................................................................................... 28

Table 12: GPC results. .................................................................................................................................... 29

Table 13: ESI-MSMS results. .......................................................................................................................... 30

Table 14: SPE-GPC results. ............................................................................................................................ 30

Table 15: Esterified UCO analysis (methods of analysis are described in 4.2.3). .......................................... 33



Abbreviations

UCO: Used Cooking Oil

GHG: Green House Gas

LCA: Life Cycle Analysis

RED: Renewable Energy Directive

FT: Fisher Tropsch

HEFA: Hydroprocessed Esters and Fatty Acids

HRJ: Hydrotreated Renewable Jet

HVO: Hydrotreated Vegetable Oil

GTL: Gas-To-Liquid

ICAO: International Civil Aviation Organization

FFA: Free Fatty Acid

PFAD: Palm Fatty Acid Distillate

n.a. : not available



Definitions

ASTM International: known until 2001 as the American Society for Testing and Materials (ASTM), is an

international standards organization that develops and publishes voluntary consensus technical standards

for a wide range of materials, products, systems, and services.

BIO-OIL: the liquid product from pyrolysis process, also called “pyrolysis oil”.

DIN: Deutsches Institut für Normung (German Institute for Standardization) is the German national

organization for standardization and is the German ISO member body. DIN is a Registered German

Association (e.V.) headquartered in Berlin. There are currently around thirty thousand DIN Standards,

covering nearly every field of technology.

DROP-IN FUEL: a fully compatible fuel for blending with traditional hydrocarbon fuels.

HYDROPROCESSING/HYDROTREATMENT: Generic term used to describe a range of refinery processes

that use hydrogen, along with an appropriate catalyst, to remove undesired components from refinery

streams. The technology is core to a modern petrochemical refinery.

RENEWABLE JET FUEL: a sustainable jet fuel derived from biomass.

UCO: Used Cooking Oil. Also referred to throughout the literature as WVO (waste vegetable oil) and UVO

(used vegetable oil), RVO (recycled vegetable oil) and RCO (Recycled Cooking Oil). UCO has been chosen

as the standard for this document



Introduction 1

The use of vegetable oils to produce renewable fuels, i.e. biofuels fully composed by hydrocarbons through

hydrotreating or FT-synthesis, consists in a real opportunity for greening the aviation sector. Defined ranges

of bio-hydrocarbons are requested to substitute the traditional jet fuel into the common commercial mixture,

according to ASTM D4054 [1] and D7566 ASTM [2] to approve the renewable jet fuel, and ASTM D1655 [3]

for Jet A-1 requirements. While FT-synthesis requires high capital cost to work the process, HEFA cost is

largely affected by the high operative cost. From the study of Pearlson et al [4], the cost of feedstock has a

relevant impact on the gate price of HEFA (60 – 70%), thus the necessity to find an alternative, lipid-based

feedstock, becomes fundamental.

The ITAKA project has the aim of substituting the traditional vegetable oils with camelina and used cooking

oil (UCO) as feedstock for commercial hydrotreating, promoting a sustainable chain for commercial aviation

transport [5]. Camelina consists in a crop cultivated in marginal lands (i.e. Spain in ITAKA), which can be

counted double towards 10% target for biofuels according to RED scheme [6] (as well as for UCO). Its

vegetable oil is suitable for commercial process of hydrotreating with no precautions than other traditional

vegetable oils. Used cooking oil is a waste that needs to be collected and recycled, and due to its nature, the

adoption of this feedstock is not possible without upgrading/cleaning pre-treatment. In order to find a strategy

to reduce the HEFA cost, ITAKA scope was focused on the investigation of alternative thermochemical

treatments to produce bio-hydrocarbons suitable for aviation fuel.

Despite UCO derives from vegetable oils, cooking process alters some relevant properties of feedstock. An

experimental campaign on filtration and centrifugation was performed and reported on D2.3 [7], focusing on

macro- and micro-contaminants reduction. The achieved results on UCO showed an important reduction of

nitrogen and phosphorous content by high quality filtration (up to 0.8 micron). Despite contaminants and

water content were strongly reduced, the still high nitrogen content could be associated to the oil soluble

proteins, which were present in the raw UCO. Moreover, acidity and other micro-contaminants such as Ca,

Mg, K, Na and other metals need to be monitored before its use as feedstock for next thermochemical

processes.

Currently there are no specifications about the use of UCO as feedstock for thermochemical processing: only

DIN 51623 [8] can provide an indication about the quality of feedstock, although it is referred to rapeseed oil

as fuel. The development of UCO specification for hydrotreating, or generally, for thermochemical processing

to produce biofuels, needs an accurate study on feedstock, process and products. In order to develop a

proposal for UCO specification, the following issues were considered:

1. the differences between traditional vegetable oils and filtered used cooking oil;

2. the interaction between used cooking oil and catalysts;

3. the presence of hydrogen during the process, as first promoter either of deoxygenation reactions, or

other reactions that involve the heteroatoms contained into the oil.

In the case of catalytic pyrolysis, which is the main subject of present document, the reaction pathways do

not require hydrogen, thus the quality of feedstock becomes less important. This work investigates the

catalytic pyrolysis of UCO, as a process to convert triglycerides in bio-hydrocarbons. The conversion route

was divided in two different steps: first the triglycerides are pyrolyzed in a fixed bed reactor; then, the



generated pyrolysis vapours are catalytically cracked in a dedicated reactor. Finally, the pyrolysis vapours

are condensed, separated by non-condensable gases, and collected as bio-oil. The main goal is based on

the development of a strategy for deoxygenating the carboxylic group of fatty acids molecules, leading the

direct production of paraffins and olefins. The investigation on the reaction pathways was carried out

experimentally, in terms of quantity and quality of collected bio-oil, by the use of different catalysts. Bio-oils

was chemically analysed by both RE-CORD and NESTE laboratories, characterizing the fuel properties and

individuating the most significant chemical species. Once the quality of feedstock and bio-oils was assessed,

a first proposal of UCO requirements as feedstock for thermochemical processes was carried out.

Figure 1: RE-CORD experimental activities on UCO.

3. UCO upgrading / cleaning pre-

treatments [D2.3]

4. Thermochemical

processing [D2.8]

5. Quality assesment on final product

(biofuel or bio-intermediate)

[D2.8].

6. . Proposal of UCO

specifications as feedstock for

thermochemical processing

[D2.8]

1. Analysis on raw UCO

batches [D2.3]

2. Individuation of critical

parameters [D2.3]



UCO as feedstock for biofuels production 2

Used cooking oil (UCO) consists in a mixture of edible vegetable oils, and sometimes animal fats, which

derived from the fried food that need large amounts of oil for cooking [9]. In general, if a vegetable oil is

heated at high temperatures (140 – 180 °C), an increase in saturation level and FFAs content occur, due to

the decomposition of the triglyceride molecules [10]. In the case of the frying process, the interaction

between oil and food cause several reaction like: hydrolysis, polymerization, oxidation [11], and in some

cases cyclization of the fatty acids [12]. As consequence, the physical and chemical properties of the

vegetable oils adopted as frying oil can significantly change; in general UCO has higher acid value, viscosity,

moisture and contamination respect to edible vegetable oils. According to literature, UCO can be defined

with several synonymous such as WCO (Waste Cooking Oil) or WVO (Waste Vegetable Oil), but in some

cases it can include a specific lipid-waste composed by animal fats (e.g. Yellow Grease).

From a legislative point of view, the Waste framework Directive 2008/98/EC1 [13] classifies UCO as “edible

oil and fat” and Directive 2004/35/CE2 [14] establishes a framework for environmental liability in order to

prevent and remedy the environmental damage, in this case by means of the regulation of UCO chain on

production/collection/recovery system. In this optic, EU Member States must control the companies

responsible for the collection and handling or treatment of UCO.

Figure 2: On the left side, raw UCO. On the right side, the residual of centrifugation (left) and the separated oil (right).

Animal By-Products Legislation 1774/20023 [15] forbids the use of UCO (originating from restaurants,

catering facilities, and kitchens) as ingredient into animal feed (or it is possible only if properly filtered and re-

esterified), so its use can only be suitable to produce biofuels and chemicals.



The use of UCO as biofuel was deeply studied, tested and currently adopted for biodiesel production [16–

19]. According to the RED scheme, UCO is considered a feedstock for biofuels produced from “wastes,

residues, non-food cellulosic material, and lignocellulosic material” which can be counted double against the

emissions reduction target of 10% renewable energy goal for transports. Currently the sub-target for the

promotion of advanced biofuels is 0.5%, without including UCO, but making its use eligible and double

counted for the 10% goal [20]. As a consequence, the market of UCO (despite it is still relatively immature)

has changed dramatically over the past few years, passing from being a waste to a product sold up to 250 –

300 €/ton for raw UCO, and up to 880 €/ton for refined oil [21]. Only few data assess the availability of UCO

for EU-27 member states: estimations on the available collectable potential range from less than 1 Mtons to

above 3.5 Mtons per year was carried out by Tacconi et al. [22]. In regard to extra UE import-export, only a

small quantity of Chinese UCO has been found to be imported to the EU. This because all actors involved in

UCO supply chain for biodiesel production, starting from UCO collector to end user, have to be certified

under one of the existing certification schemes (ISCC EU and ISCC DE) from January 1st, 2013 [23]. This

fact is going to limit the import of UCO in EU, in particular for biodiesel production (UCOME), thus the market

share from other countries is going through a marked reduction. Currently, approximately 80 - 90% of the

UCO collected in the EU-27 is used for biodiesel production, and the other fraction for energy and

oleochemicals [24]. Just a small quantity is dedicated to other use, out from fuels and chemical market.

2.1 UCO vs traditional vegetable oils

The properties of UCO as biofuel was deeply investigated in D2.3 “Report describing the main barriers to

innovation and recommendations for further research” [7]. Samples of UCO and commercial sunflower oil

were analysed to investigate how the cooking process affects the vegetable oils properties as biofuels. From

the analysis of several batches of UCO, it emerges higher acidity due to higher free fatty acids content,

higher contamination and water content, higher micro-contaminants content (N, P, S, Ca, Mg) than traditional

vegetable oils. The adoption of UCO as feedstock in hydrotreating was not possible without

upgrading/cleaning pre-treatments. In order to develop a feedstock suitable for the existing infrastructure, an

experimental campaign on filtration and centrifugation was performed to remove macro- and micro-

contaminants. The achieved results on UCO showed a relevant reduction of contaminations, water content,

nitrogen and phosphorous content with a high quality paper filtration (up to 0.7 micron pore size). Despite the

achieved improvements, the UCO nitrogen content of still remains a significant amount (80 ppm).

2.2 Industrial esterification of UCO

The main use of UCO as feedstock for biofuels production consist in the conversion in biodiesel. Biodiesel

(FAME, Fatty Acid Methyl-Ester) represents the most traditional biofuel in the EU, produced by

transesterification of triglycerides. In order to produce 3 moles of biodiesel and 3 of glycerol, 3 moles of

alcohol are required to react with one mole of triglyceride [74]. The reaction includes catalysts: in commercial

treatment are commonly adopted alkaline solutions, such as sodium or potassium hydroxide, to accelerate

the reaction [75]. In contrast, the presence of free fatty acids in UCO inhibits the transesterification reaction

[25]. An industrial process to restore the original triglycerides content of UCO consists in the esterification

process. A local company of the Florence district, SILO SpA, works on exhausted cooking oil and fats from



several years, collecting the feedstock by means of contracts with private companies. After collection, SILO

leaves the UCO to sediment and then filters it at about 50 micron. A part of UCO is then subjected to

esterification process, carried out at about 220 - 230 °C, with the addition of glycerol (pure glycerine at 97 –

98 wt%). This process restores the acidity of UCO to comparable values with pure vegetable oils. Three

batches of esterified UCO were sampled and analysed. Table 1 shows the properties of esterified UCO, the

same as those investigated for raw and filtered UCO in D2.3, and compared to DIN 51623 [8] “Fuels for

vegetable oil compatible combustion engines” (a recent German standard for pure plants oil).

Table 1: Esterified UCO analysis (methods of analysis are described in 4.2.3).

Parameter Unit Method DIN 51623 UCO EST 120126 UCO EST 130528 UCO EST 130704

Density Kg m-3 UNI EN ISO 3675 900 - 930 938 916 922

Kinematic Viscosity

at 40 °C mm

2 s

-1 UNI EN ISO 3104

< 35

34.5 34.08

Acid value mgKOH g-1 UNI EN 14104 < 4 0.323 2.4 3.3

Free Fatty Acid wt% UNI EN ISO 660 < 2 0.1615 1.2 1.65

Iodine Number g Iodine 100g-1 UNI EN 14111 < 125 86 96 96

Water content wt% UNI EN ISO 8534 < 0.075 0.016 0.944 2.1

Ash wt% UNI EN ISO 6245 - 0.01 0.03 0.04

Total contamination mg kg-1 UNI EN 12662 < 24

628 1,895

Sulphur wt% Internal method < 0.001 0.136 0.002 0.014

Calcium mg kg-1 UNI EN 14538 < 1 15.5 0 0

Magnesium mg kg-1 UNI EN 14538 < 1 2.7 0. 0

Phosphorus mg kg-1 UNI EN 15290 < 3

4.933 7.5

C wt% ASTM D5291 - 73.5 76.5 75.8

H wt% ASTM D5291 - 11.2 12 11.8

N wt% ASTM D5291 - 0 0.029 0.028

O wt% calculated - 15.164 11.439 12.318

Calorific value, higher kJ kg-1 DIN 51900-2 > 36,000 37,898 38,729 38,225

Calorific value, lower kJ kg-1 calculated - 35,455 36,183 35,721

Despite the free fatty acids content was fully restored to the acceptable limits of traditional vegetable oils, the

esterified UCOs show unacceptable water content, ash content, micro- and macro-contaminants, which were

strongly limited by high quality filtration in D2.3. Next chapter investigates alternative lipids-to-biofuels

pathways than transesterification, such as hydrotreating and catalytic pyrolysis: these processes can work

with the presence of free fatty acids because they are focused on direct de-oxygenation of triglycerides and

fatty acids molecules, where the oxygen is located (i.e. carboxyl group). So the esterification process of UCO

is useful only to improve the transesterification conversion yield, favouring the production of an ester

compound like biodiesel. Considering the direct conversion of triglycerides into bio-hydrocarbons, the

presence of free fatty acids in UCO does not affect the conversion process, thus the esterification of UCO as

upgrading process was not considered.



Lipids to biofuels pathways 3

One of the main challenge in biofuels production consists in the best grade of deoxygenation from initial

biomass to final product. Lipid-based feedstock consist in high quality material for biofuels production due to

its low oxygen content compared to other type of biomass (such as lignocellulosic material, which are mainly

composed by sugars and polymers). Oxygen content ranges from 39 – 42.5 % [26] for wood, to 10 - 12% of

vegetable oils [27] (where these values are in mass, dry and ash free basis). Moreover, the oxygen bonds

into chemical structure of lipids are easier to separate than complex polymeric structure of sugars, where the

oxygen is located. Lower oxygen content means higher energy density (i.e. higher C-H content), so the

challenge consists in the deoxygenation of carboxyl group of triglycerides (i.e. major constituent of lipids).

There are three methods to crack long chain of hydrocarbons to short chain of hydrocarbons [28].

1. The first method is known as thermal cracking and occurs with the aid of heat to produce a lighter

product.

2. The second process is the catalytic cracking that is conducted in the presence of an acid catalyst

without the use of hydrogen and needs less thermal energy than thermal cracking.

3. The third method is known as catalytic hydrocracking (or hydrotreating) and occurs in the presence

of a bifunctional catalyst and a high hydrogen pressure. The catalytic hydrocracking process

consumes less thermal energy and the presence of hydrogen minimizes coke formation and

therefore reduces catalyst deactivation by pore blockage. However the cost of hydrogen have a

relevant impact on the economic costs.

A scheme with inputs (lipid-based feedstock), processes, and products (biofuels) were reported in Figure 3.

The main technology able to remove the heteroatoms of oxygen from biomass is the derived technology from

oil refineries named “hydrotreating”. The product from lipids hydroprocessing, HVO (Hydrotreated Vegetable

Oil), consists in a renewable fuel with the same characteristics of fossil fuel (i.e. full aliphatic hydrocarbons

fuel). Despite its high fuel quality, the process requires relevant operative and maintenance costs, thus in

recent years the research looked towards novel processes based on thermochemical conversion, mainly

catalytic pyrolysis (based on catalytic cracking). In this section a commercial and an experimental pathway to

produce renewable hydrocarbons are evaluated.



Figure 3: Thermochemical conversion routes from lipids to biofuels [29].

3.1 Hydroprocessing of vegetable oils

Hydroprocessing is used in the petroleum refinery to remove S, N and metals from petroleum-derived

feedstock, including heavy gas–oil or vacuum gas–oil. These processes, even into existing refineries, can be

used for lipids hydroprocessing, introducing the biorefinery concept. These treatments can be sometimes

called “hydrotreating” when applied to oxygenated paraffin chains, such as triglycerides or FFAs, as they

lead to the removal of oxygen by hydrogenation, decarboxylation and decarbonylation [30]. A scheme of the

ideal reactions pathway was carried out by Huber et al. [31] and reported in Figure 4, showing the production

of iso-paraffins in the range of diesel fuel. The reaction conditions, together with catalyst nature, are crucial

to define the performance and yield of conversion. Sotelo-Boyás et al. [28] reported a detailed review on type

of oil, reactor, reaction conditions, catalyst, and main products obtained during hydroprocessing of pure oil

and oil plus gas oil mixture.

Figure 4: Scheme of the main reactions in the hydrotreatment of generic vegetable oils and fatty acids [31].

Since few years, several companies such as Neste Oil, Petrobras and ENI/UOP started the commercial

production of HVO, both for road and air transport fuels [32]. A recent study found that the baseline cost for



HEFA fuel production ranges between 1.01 - 1.16 $/liter, where the cost of process impact for 30 – 40 % of

total cost per litre [4]. Considering the all operating and under commissioning World plants, currently the

global HVO/co-processing production reaches 4.9 Mtoe/yr, where 2.8 from EU sites only [6].

Figure 5: Neste Oil’s renewable diesel plant in Singapore, with a capacity of 800,000 tons per year. Source: neste.com

3.2 Experimental pathways

While biodiesel is a traditional biofuel with defined blend walls (up to 20% in diesel engines without requiring

any modification [33]), currently the fuel demand is going towards the production of renewable hydrocarbons

due to their ability to be drop-in fuels. In addition to hydrotreating, a possible alternative pathway consists in

the adoption of the current technology of petroleum refineries, mixing lipids with petroleum fractions such as

straight run gas oil and/or vacuum gas oil, in order to be processed on conventional hydrotreating reactors

[28]. Another alterative consists in the hydrothermal conversion of vegetable oil by means of batch reactors

in severe reaction conditions [34], but also this way implies high operative energy costs. The next paragraph

explores the catalytic pyrolysis of lipid-based feedstock: the simplicity of process and the promising

experimental results from literature make of this process a promising alternative for the production of

renewable hydrocarbons.

3.2.1 Catalytic pyrolysis of lipids

Thermal and catalytic pyrolysis of lipid-based feedstock represents a pathway for the production of

renewable bio-based products such as fuels and chemicals, with a final products composition similar to

diesel fuel [35]. Several works on lipids reaction schemes were proposed [36–40]. The cleavage of the

triglycerides molecules originates a mixture of hydrocarbons of smaller chains and oxygenated compounds,

such as alkanes, alkenes, alkadienes, aromatics, aldehydes, ketones, and carboxylic acids. Teixeira and

Sousa [41] proposed a mechanism based on two step of cracking that occur simultaneously: first (i.e.

primary cracking), acid species (mainly carboxylic acids) are formed during the thermal decomposition of

triglycerides by means of the breakdown of C-O bonds of carboxylic group, then (i.e. secondary cracking) the

products from the first reaction lead the formation of organic compounds with shorter chain lengths, including

saturated and/or unsaturated hydrocarbons. The effect of catalysts affects the selectivity of secondary

reactions, e.g. in the case of zeolite cracking favours the formation of aromatics compounds by means of

complex reaction mechanisms. Given the difficulties to understand a defined reaction mechanism, Zhenyi et

al. [42] studied pyrolysis mechanism of triglycerides focused on triolein thermal degradation (selected as

reference structure for vegetable oils). Calculation results gave that the cleavage of C−O bond takes place at

the temperature of about 288 °C while the C-C bond at β position occurs at about 388 – 426 °C. However,



the initial pyrolysis (at temperature above 400°C) may begin with the contemporary cleavage of C−O bonds

and β bonds of double carbon bonds (C=C). In general the presence of unsaturation favours the cracking in

proximity of the C=C bonds [43], and the energy of the C–C bonds in n-alkanes decreases by moving

towards the centre of the molecule. Currently the exact reaction mechanism of thermal decomposition of

lipids remains a challenge due to the multiple cracking reactions and re-polymerization that occur during

pyrolysis. A true indication of the nature of reaction products is given by experimental tests investigated in

next chapter.

Figure 6: Reaction scheme for the pyrolysis of saturated and unsaturated triglycerides proposed by Idem et al [39].



Experimental campaign on catalytic pyrolysis of UCO 4

4.1 Goal and Scope

Catalytic pyrolysis could be effectively applied to lipids and fatty acids to remove the oxygen and generate

hydrocarbons. The potential benefit of pyrolysis versus hydroprocessing relies in the fact that the process

does not require hydrogen and noble catalysts, or alternatively can operate with low-cost catalysts, thus

more tolerant to irregularly contaminated feeding as UCO. Moreover, pyrolysis can be better downscaled

than hydroprocessing, and thus it – especially targeting road transport fuels - could be implemented at

smaller scale in decentralised unit, placed close to the collection points of residual oils and well integrated in

a local circular economy scheme. Finally, catalytic pyrolysis can be evaluate as first step of UCO conversion

before hydrotreating, to save hydrogen demand.

The present process of catalytic pyrolysis of used cooking oil is based on two separate thermochemical

steps of conversion: a first primary thermal cracking of UCO; a secondary catalytic upgrading (selective

cracking) of generated pyrolysis vapours from first step of process. The main goal is based on the

development of a strategy for deoxygenating the carboxylic group of fatty acids molecules (i.e. principal

constituents of triglycerides), leading the direct production of bio-hydrocarbons (in the range of aviation

fuels). The target bio-hydrocarbons suitable for aviation fuel, as well as the characteristics of commercial and

renewable aviation fuels, were investigated in D2.3. The investigation on the pyrolysis products was carried

out experimentally, in terms of quantity and quality of collected bio-oil. Four catalysts were tested at different

process conditions. In order to evaluate the effect of pyrolysis on an alternative lipid-based feedstock, a final

test on pure unsaturated fatty acids was carried out, and the liquid product was compared to UCO pyrolysis

oil.

Figure 7: A scheme for RE-CORD catalytic pyrolysis of triglycerides-based materials.

4.2 Materials and methods

4.2.1 Pyrolysis unit

A dedicated continuous catalytic pyrolysis unit was designed, assembled and adopted in the framework of

ITAKA. The unit can process 1.5 kg h-1

of bioliquids. The experimental rig is composed by a first pyrolysis

fixed bed reactor, able to convert the lipids into pyrolysis vapours, that are subsequently upgraded by a

second dedicated catalytic reactor, and finally condensed into a bio-oil (pyrolysis oil). The feeding system

consists in a pressurized injection system, electrically heated, to reduce the oil viscosity and improve

atomization in the reactor. The pyrolysis chamber is heated by 12 kWe electrical resistors, able to maintain



the operative temperature up to 550 °C: the reactor temperature is measured and controlled by means of a K

type thermocouple place in the middle of the reactor. The catalytic reactor, properly heated by a dedicated

resistor (3 kWe), up to 550 °C, allow the catalytic upgrading of pyrolysis vapours. During operation, a nitrogen

flow is heated and flushed into the thermal reactor at a constant flow rate of 4 L min-1

in order to maintain the

inert atmosphere. The condensation system consists of two water-cooled condensers (water at 15-20 °C)

connected in series and a water bubbler to remove the aerosol from non-condensable gases. Pyrolysis

gases enter the first condensation section at 400°C, leave this first condenser at ≈80 °C, and exit at ≈20°C

from the second condenser. Pyrolysis oil from the two condenser is mixed and collected in a bottled placed

at the bottom of the two condensers. The water bubbler temperature is kept at 15-20 °C. Permanent gases

are sucked through the whole plant and discharged by a blower. A schematic representation of the process

is given in Figure 4. Char is batch-wise discharged at the outlet of the pyrolysis chamber. Liquid yield was

calculated as the ratio of the mass of the collected pyrolysis oil to the mass of processed feedstock.

Figure 8: A representative scheme of the catalytic pyrolysis unit for UCO conversion.

4.2.2 Feedstock

UCO was supplied by SILO: the company acquires raw UCO from restaurants and eating-places in the

metropolitan area of Florence. The oil is then stored in dedicated stainless steel tanks at SILO premises. The

collecting of samples from these UCO tanks requires the use of a special tool, named “sonda Marsigliese”,

which is also used as sampling methods for the acceptance of each UCO batch. This instrument allows

taking UCO samples at the same time from different heights in the tanks, so to ensure that the collected

liquid is representative of the whole liquid stored in the tank. UCO is then pumped through 1 μm mesh

cardboard filters, that reduce contamination as well as water content (MP Filtri GRF015 portable off line oil

transfer/filtration unit). This filtered UCO was adopted as feedstock for the present work. The Fatty Acid (FA)

mixture was instead supplied by InnovHub (Milan, Italy), formerly known as Experimental Station for Oil and

Fats.



4.2.3 Methods of analysis

Bio-liquids were characterized before and after the catalytic pyrolysis process, according to the following

procedures.

i. Elemental analysis

The elemental analysis was performed by LECO-Truespec according to ASTM D5291 [44]. A 50-100 mg

sample was used to perform the determination of C, H, N, while oxygen content was obtained by

difference. The TruSpec Add-On Module was used for the determination of Sulphur content according to

an internal method. The more accurate nitrogen content was measured by Thermo Finnigan FlashEA

1112 NC analyzer, according to ASTM D4629 [45].

ii. LHV and HHV determination

HHV was determined by a LECO-AC500 calorimeter following DIN51900-2 [46]. LHV was calculated

from HHV taking into account moisture and Hydrogen contents as specified in DIN51900-1 [47].

iii. Acid value and free fatty acid content

The acid value, used to assess the free fatty acid concentration in the feedstock, was evaluated as the

amount (milligrams) of potassium hydroxide necessary to neutralize a mixture of the sample blended

with a proper solvent (Ethanol: Diethyl ether 1:1). According to the ISO 660 [48], a manual titration is

used as reference method for this analysis. The determination of free fatty acid content was verified

according to EN 14104 [49]. The acid value is expressed as a mass fraction, the free fatty acid content is

expressed as a percentage mass fraction.

iv. Kinematic Viscosity

Kinematic viscosity of feedstock and pyrolysis products was determined at 40°C by a LAUDA Proline

PV15 viscometer, according to the reference norm UNI-EN ISO 3104 [50].

v. Water content

The water content was measured by Karl Fisher’s titration according with the UNI 8534 [51] for the

feedstocks. For the pyrolysis products, the water content is determined with the modified procedure

proposed by Oasmaa and Peacocke [52]

vi. Macro and microelement determination

Macroelements (such as Ca, Mg and P) were measured according to UNI EN 14538 [53] and UNI EN

15290 [54] by means of ICP, IRIS Intrepid II Thermo Fisher.

vii. Total contamination

Total contamination was measured according to UNI 12662 [55], adopting a high retention glass-fibre

filter of 0.7 μm (mean pore size), solubilizing the sample in n-heptane.



viii. Density

Density of liquids was measured at 15°C by a calibrated glass hydrometer according to UNI 3675 [56].

ix. Ash

Ash was determined with a Leco TGA 701 according to EN ISO 6245 by heating up to 750 °C and

holding there for at least 1h 30min.

x. Insoluble impurities

Insoluble impurities was measured according to EN ISO 663 [57], adopting a Glass-fibre filtre (Whatman

GF/D type, i.e. 2.7 μm mean pore size), solubizing the sample in n-exane.

xi. Iodine number

Iodine number was measured according to procedures, reagents and calculations indicated in EN 14111

[58].

xii. Determination of fatty acid composition by GC-FID technique

The fatty acid composition of feedstock was determined by GC-FID GC 2010 Plus (Shimadzu) equipped

with a HP88 column (Agilent) (length 60 m, internal diameter 0.250 mm, film diameter 0.20 μm).

Transesterification was performed according to EN ISO 5509 [59]. The analysis was carried out with a

column flow of 0.46 cm3 min

-1 at initial temperature of 120°C (holding time: 1 min) increased to 175°C

(heating rate 10 min, holding time 10 min), 210°C (heating rate 5 min, holding time 28 min) and finally

230°C (heating rate 5 min, holding time 28 min). The identification of different methyl ester peaks was

carried out by comparison with the retention time of the pure compounds (FAME Sigma Aldrich mix

standard).

xiii. Chemical class composition by gas chromatography techniques

The composition of hydrocarbons, fatty acids and other compounds in the pyrolysis oils was measured

using GC-MS and GC-FID (GC 2010 Plus-Shimadzu). Both analyses were performed using 500 mgr of

sample, dissolved in 25 cm3 of isopropanol and then injected (1 mm

3) in split mode; the used column

was a Zebron ZB-5HT INFERNO (Phenomenex) (length 30 m, internal diameter 0.250 mm, film diameter

0.25 μm). The analysis was performed with a column flow of 2.02 cm3 min

-1 for GC- MS and 3.17 cm

3

min-1

in GC- FID with an initial temperature of 40°C (holding time 10 min) increased to 200°C (heating

rate 8 min, holding time 10 min) and then to 280°C (heating rate 10 min, holding time 30 min). For GC-

MS the mass to charge ratio (m/z) ranged from 35 to 550. Compounds were first identified in GC-MS by

the comparison of their mass spectra with the NIST library. For quantification of specific chemical

compounds in GC-FID, their response factors, relative to the internal standards o-terphenyl or D-6

benzene, were determined by using standard compounds injected in a range of concentrations that

reflects those found in the samples. All standards are provided by Sigma Aldrich. Regarding the

hydrocarbons standards the compound considered were n-paraffins, n-olefins, aromatics (BTEX and

PAHs), As chromatograms of catalytic intermediate pyrolysis of lipid-based materials show a hump at the



central zone (at about 15-30 min), similar to diesel/gasoline chromatograms, even if less accentuated, in

order to maintain a conservative approach as regards quantitative assessments of hydrocarbons in the

pyrolysis oil mixture, a “valley-to-valley” integration has been performed, thus not considering the bottom

area. This approach, while probably underestimating the actual amount of hydrocarbons, avoids possible

interferences of unidentified chemicals with the pure peaks in the chromatogram and therefore

overestimation of results.

xiv. Distillation procedure

Two samples of pyrolysis oil were analysed for a detailed investigation by separating their boiling

fractions. The distillation was carried out at atmospheric pressure by a distillation rig heated with a

diathermic oil bath, using a 1 kW heating plate. Distillation temperatures were selected on the basis of

the most representative boiling ranges of paraffinic hydrocarbons fuels, and three different fractions were

separated: A (< 150°C); B (150 – 200 °C); C (200 – 250 °C).

4.2.4 Catalysts

Four catalysts were selected for present experimental campaign. Catalysts were bought by commercial

companies, in form of pellets or spheres. The selection criteria on catalysts type was based on a review

reported in precedent works [60–62]. In order to screen the best catalyst performance, a neutral, a moderate

and strong acid, and a strong basic catalyst was selected. Catalysts features are shown in Table 2. Before

thermochemical processing, each catalyst was maintained into the thermal reactor at 500 °C for 2 hours,

under a nitrogen flow to maintain the inert atmosphere.

Table 2: Evaluation of properties of the adopted catalysts.

Product name Catalyst Shape Dimensions Acidity Si/Al

ratio

Surface

area Bulk density

- - - Diameter x

Length [mm] - - [m

2 g

-1] [kg dm

-3]

Gale Galicarb

ESO 4

Activated

carbon pellet 4 x 2-10 Neutral

0.45

Gale GAL 501 Alumina spheres

6-8

(diameter

only)

Moderate

acid 300 0.80

ACS Material H-

ZSM-5 Zeolite pellet 2 x 2-10

Strong

acid 38 - 50 > 300 0.72

Tritrust MgO

99%

Magnesium

oxide pellet 3 x 6

Strong

basic

4.2.5 Experimental runs

A summary of the experimental tests is reported in Table 3. First tests were initially carried out without

catalyst in order to evaluate the best conversion temperature in terms of bio-oil yield at 500 °C temperature.

The first catalytic run of experiments investigated the effect of four selected catalysts on pyrolysis products,

at two different catalytic reactor temperatures (i.e. 500 and 550 °C) and fixed WHSV (mass feed-to-catalyst



weight ratio) equal to 4 h-1

. The catalyst mass was then increased to 2.5 h-1

WHSV in following tests with the

best performing catalyst (i.e. activated carbons). A final test was carried out with pure fatty acids and with the

same configuration.

Table 3: Description of experimental activities.

Activity Description Feedstock Test

N. Catalyst

Thermal

reactor

temperature

Catalytic

reactor

temperature

WHSV

n - - n - °C h-1

1 First experimental

campaign UCO 1 none 500 500 -

1.1.1. UCO 2

Activated

carbons 500 500, 550 4

1.1.2. UCO 3 Alumina 500 500, 550 4

1.1.3. UCO 4 H-ZSM-5 500 500, 550 4

1.1.4. UCO 5

Magnesium

Oxyde 500 500, 550 4

2

Second

experimental

campaign

UCO 6 Activated

carbons 500 500 2.5

1.1.5. FAs 7

Activated

carbons 500 500 2.5

4.3 Results

This section reports the experimental results from the catalytic pyrolysis campaign and feedstock and

products characterization.

4.3.1 Oil-based feedstock

Filtered UCO and pure FAs have been characterized in most of relevant biofuel properties. After high grade

filtration (up to 1 micron), the UCO presents similar characteristics of a traditional vegetable oil (in this case a

commercial sunflower oil, reported in Table 4). A batch of commercial sunflower oil was compared to filtered

UCO and pure FAs. UCO viscosity is slightly higher than sunflower oil, while the water content and the

contaminations are strongly reduced (by high grade filtration) from the original feedstock (analysed in D2.3)

to comparable values of traditional vegetable oils. Ca, Mg and P content are even under DIN 51623, and

there are no other significant changes compared to traditional sunflower oil. A relevant difference is observed

on the acid values: while UCO and vegetable oils may contain a small fraction of free fatty acids, pure fatty

acids are in the range of about 200 mgKOH g-1

(for vegetable oils the fatty acid content can be estimated

equal to half than the acid value, according to Knothe and Steidley [10]).



Table 4: Analytical analysis on feedstock .

Parameter Unit Method Sunflower oil UCO filtered at 1 μm Pure FAs

Density kg m-3

UNI EN ISO 3675 920 911

Kinematic Viscosity at 40 °C mm2 s

-1 UNI EN ISO 3104 27.84 38.15 18.03

Acid value mgKOH g-1

UNI EN 14104 7.2 2.63 201.7

Free Fatty Acid wt% ISO 660 3.6 1.31 100.85

Iodine Number gIodine 100g-1

UNI EN 14111 118 93

Water content wt% UNI EN ISO 8534 0.125 0.080 0.08

Ash wt% UNI EN ISO 6245 0 0.01

Total contamination mg kg-1

UNI EN 12662 206.3 256 68

Insoluble impurities wt% ISO 663 0.01 0.05

Sulphur mg kg-1

Internal method 52 0

Calcium mg kg-1

UNI EN 14538 39.51 0

Magnesium mg kg-1

UNI EN 14538 3.17 0.22

Phosphorus mg kg-1

UNI EN 15290 8.075 10.1

C wt% ASTM D 5291 76.5 76.3 76.3

H wt% ASTM D 5291 11.3 11.7 12.2

N wt% ASTM D 5291 0.33 0.02 0.02

N mg kg-1

ASTM D 4629 116 137

O wt% Calculated 11.87 11.98 11.48

Calorific value, higher MJ kg-1

DIN 51900-2 39.32 38.91 39.4

Calorific value, lower MJ kg-1

Calculated 36.92 36.43 36.8

Fatty acids profile of both feedstock were investigated by GC techniques. While for UCO the analysis

represents a traditional fatty acid profiles as well as vegetable oils, for pure fatty acids it consists on a full-

composition analysis due to absence of glycerol. As regards Table 5, UCO is mainly composed by palmitic,

oleic and linoleic acid (15, 43 and 34 wt% respectively), while pure fatty acid are mainly composed by oleic

acid (i.e. about 62 wt%) and small fractions of other fatty acids, due to its olive oil original nature.

Table 5: Fatty acids profile of adopted feedstock.

Fatty acids Chemical structure UCO Pure FAs

- - wt% wt%

Lauric C12:0 0.09 2.83

Mystiric C14:0 0.40 1.43

Palmitic C16:0 15.11 4.03

Stearic C18:0 3.96 1.17

Oleic C18:1 43.43 61.81

Linoleic C18:2 33.93 13.47

Linolenic C18:3 0.34 0.38

Erucic C22:1 0.00 6.41

Others - 2.75 8.47

4.3.2 Results from first experimental campaign

The process temperature for the first step of pyrolysis was selected by precedent tests. The temperature of

500 °C was selected as the most performant in terms of bio-oil yield (i.e. empty reactor test). Once the

thermal reactor temperature was fixed, eight tests (four catalysts, tested at 500 and 550 °C into catalytic

reactor) were performed. Tests conducted at 550 °C led to the formation of higher fractions of non-

condensable gases against lower fraction of bio-oil, thus the following investigation of products was carried

out at constant 500 °C in both pyrolysis steps.



Table 6: First run yields of catalytic campaign at 500 °C.

Depending on the catalyst type, products present different characteristics, in terms of products yields (Table

6) as well as in chemical and physical properties of bio-oils. Bio-oils are collected in the range from 33.74 to

63.64 wt%, depending on catalyst, and the unconverted fraction collected at the end of each test is about 15

wt% of total feed (because catalysts aging on pyrolysis vapours). This leads to estimate the permanent gas

yield by difference, that is minimized in test with activated carbons (21.36 wt%), favouring the maximization

of bio-oil collected. The characteristics of bio-oils are shown in Table 7. From elemental analysis emerges

that the overall deoxygenation grade is lower than expected, but significant changes are observed in the

chemical composition as shown in GC analysis. Water content is higher in test with zeolite, where the

produced reaction water is probably caused by the higher grade of cracking, which led the production of

larger amount of permanent gases. Density is reduced up to 0.85 kg l-1

in all tests, and more significantly the

viscosity values were reduced in the range of 5.68 - 1.19 cSt, close to fossil fuel parameters (e.g. 0.845 kg l-1

and 4.5 cSt as density and viscosity limits for road Diesel fuel according to UNI EN 590:2010). Acid value is

strongly increased due to FFAs content and other acidic compounds derived from catalytic cracking of

triglycerides molecules (probably carboxylic acids). The catalytic effect produced high values of acidity in bio-

oils, ranging from 117.73 (test without catalyst) to 20.45 (test with H-ZSM-5). Finally the measured heating

value is slightly higher than original feedstock despite the CHNO content was approximately maintained

constant.

0% 10% 20% 30% 40% 50% 60% 70%

Empty reactor

Act. Carbons

Alumina

H-ZSM-5

MgO

Yield wt%

Cat

alys

ts

Empty reactor Act. Carbons Alumina H-ZSM-5 MgO

Liquid Yield 62.70% 63.64% 54.55% 33.74% 61.72%

Permanent gases Yield 22.30% 21.36% 30.45% 51.26% 23.28%

Not-converted fraction 15.00% 15.00% 15.00% 15.00% 15.00%

500 °C of Thermal and Catalytic Temperature, WHSV = 4 h-1



Table 7: Analysis on bio-oil quality for each catalytic test.

Parameter Unit Norm Empty reactor

Act. Carbons

Alumina H-ZSM-5 MgO

C wt% UNI 15104 76.75 76.10 76.03 77.90 78.17

H wt% UNI 15104 11.45 11.10 11.65 10.10 11.48

N wt% UNI 15104 0.02 0.03 0.08 0.07 0.03

O wt% calculated 11.98 12.77 12.25 11.93 10.33

Water content wt% UNI 8534 0.89 0.77 1.18 8.36 1.61

Acid value mgKOH g-

1

UNI 14104 117.73 74.10 61.97 20.45 80.07

Density Kg dm-3

UNI 3675 0.87 0.85 0.85 0.90 0.86

Kinematic viscosity (40°C)

cSt UNI 3104 4.47 2.24 3.36 1.19 5.68

LHV MJ kg-1

calculated 36.82 37.79 39.57 36.48 37.46

HHV MJ kg-1

DIN 51900-2

39.25 40.14 42.02 38.55 39.95

A detailed identification and quantification of the chemical species contained into bio-oil has been carried out

following the procedure in Paragraph 4.2.3 and reported in Table 8. Essentially, this table shows the most

significant chemical species for biofuels contained into bio-oil, in particular the hydrocarbons fraction and

FFAs produced for each test.

Table 8: Composition of bio-oil fractions on first run tests analyzed by GC MS/FID.

In this case, the GC-MS analysis is adopted to identify the chemical species and then, the GC-FID is aimed

to quantify the fraction of detected compounds (mainly hydrocarbons and FFAs) which are identified by

means of internal standards of the instrument (Shimadzu GC 2010 plus). A large amount of unidentified

classes such as oxygenated compounds (e.g. ketones, aldehydes, carboxylic acids and esters [35]), other

free radicals from cracking reactions and polymerized products are not detectable by this method. The

Fatty Acids Paraffins Olefines Aromatics Ciclo-octeneUndetected

Species

Empty reactor 40% 3% 6% 2% 2% 48%

Act. Carbons 13% 5% 8% 8% 2% 64%

Alumina 12% 3% 7% 1% 2% 75%

H-ZSM-5 3% 1% 0% 4% 0% 92%

Magnesium Oxide 18% 1% 3% 1% 2% 75%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

We

igh

t p

erc

en

tage

(w

t%)



interest mainly goes towards hydrocarbons classes such as paraffins, olefins and aromatics, and their

distribution into each sample of bio-oil. Moreover, the FFAs content indicates the conversion grade obtained

by the catalytic upgrading of vapours due to the effect of catalysts, that favours secondary reactions (e.g.,

decarboxylation, decarbonylation, aromatization, cyclization, hydrogenation) according to scheme proposed

by Idem at al. [39] for the catalytic conversion of canola oil. From non-catalytic reference test, the FFAs

content into bio-oil is 40 wt%, which has been strongly reduced after each catalytic tests. The test with H-

ZSM-5 (i.e. 3 wt%) as catalyst the 92 wt% as the sum of undetected species, maybe due to high grade of

cracking that leads the production of a large amount of undetectable species. Other catalytic tests show

promising results in terms of hydrocarbon content, in particular the bio-oil from activated carbons test

contains the 23 wt% as the sum of detected hydrocarbons, which is the highest. Except the test with zeolite,

all tests indicate the presence of 2 wt% of ciclo-octene. From these results, activated carbons test was

selected for a following run at WHSV = 2.5 h-1

, i.e. increasing the catalyst mass and maintaining the other

process parameters unchanged.

Figure 9: (a) back view of the reactor and condensation system during experimental campaign; (b) bioliquid feeding

system during operation.

4.3.3 Results from second experimental campaign

The identification of the best catalyst performance was deeply investigated by increasing the catalyst mass

up to WHSV = 2.5 h-1

(i.e. test with activated carbons, AC*), maintaining the same temperature and the other

parameters of last experiments. Two tests were carried out: first maintaining the filtered used cooking oil as

feedstock, and then testing the most performant catalytic effect on pure fatty acids. In terms of products

yields, used cooking oil test with increased activated carbons mass (UCO AC*, Table 9) achieved the same



products quantities as the previous test, while pure fatty acids processing converts about 50 wt% into bio-oil

fraction (non-converted fraction and permanent gases yields were not measured).

Table 9: Second run yields of catalytic campaign. UCO AC test consists in precedent test reported in Table 6. Asterisk

(*) means tests carried out with increased catalysts mass.

Significant results were obtained by elemental analysis as shown in Table 10. Despite the oxygen

content of UCO AC run is approximately maintained constant, compared to previous tests, relevant

differences are observed in terms of produced hydrocarbon fractions. No other significant changes are

observed except a reduced acid value. Relevant results are obtained in pure fatty acid test: from the original

feedstock to bio-oil, the deoxygenation grade reached the 67.5 wt% (from 11.48 to 3.73 wt% of oxygen

content) and consequently, viscosity is reduced and heating value considerably increased. The importance

of oxygen reduction corresponds to higher hydrocarbons fractions detected by GC analysis.

Table 10: Analysis on bio-oil quality for catalytic tests at reduced WHSV. UCO AC test consists in precedent test

reported in Table 6. Star (*) means tests carried out with increased catalysts mass.

Parameter Unit Norm UCO AC

UCO AC*

FAs AC*

C wt% UNI 15104 76.1 76.3 83.53

H wt% UNI 15104 11.1 11.50 12.7

N wt% UNI 15104 0.03 0.04 0.04

O wt% calculated 12.77 12.17 3.73

Water content wt% UNI 8534 0.77 0.54 0.17

Acid value mgKOH g-1

UNI 14104 74.1 51.45 44.4

Density Kg dm-3

UNI 3675 0.85 0.84

Kinematic viscosity (40°C)

cSt UNI 3104 2.24 2.48 1.5

LHV MJ kg-1

calculated 37.79 38.9 40.1

HHV MJ kg-1

DIN 51900-2

40.14 41.34 42.79

0% 10% 20% 30% 40% 50% 60% 70%

UCO AC

UCO AC*

Fatty Acids AC*

Yield wt%

Fee

dst

ock

/Cat

alys

t

UCO AC UCO AC* Fatty Acids AC*

Liquid Yield 63.64% 63.41% 48.82%

Permanent gases Yield 21.36% 21.59%

Non-converted fraction 15.00% 15.00%

500 °C of Thermal and Catalytic Temperature, WHSV = 2.5(*) or 4 h-1



The following table shows the results of GC analysis, which also compared with previous test corresponding

results.

Table 11: Composition of bio-oil fractions on second run tests analyzed by GC MS/FID. UCO AC test consists in

precedent test reported in Table 6. Asterisk (*) means tests carried out with increased catalysts mass.

Increasing the catalyst mass on UCO test, the FFAs content is reduced in favour of higher paraffinic and

olefinic hydrocarbons fraction, respectively from 5 to 15 wt% and from 8 to 11 wt%. However, the aromatic

fraction is reduced from 11 to 8 wt%, while the ciclo-octene content is slightly increased up to 3 wt%. It

appears that larger paraffins and olefins content into bio-oil is favoured by decarboxylation and

decarbonylation reactions occurred by longer residence times of the hot vapours into the catalyst reactor.

Despite the lower liquid yield, pure fatty acid test results in higher paraffins and olefins fractions, with a

higher overall detected hydrocarbons content (from 35 wt% for UCO AC* to 40 wt% for FAs AC*). In this

case their higher fraction of total hydrocarbons is probably caused by the easier removal of carboxylic group

from fatty acids molecules, due to absence of glycerol into feedstock.

Figure 10 shows the UCO AC* spectrum from GC FID analysis, that identifies the nature of hydrocarbon

classes. In particular, higher peaks consist in paraffinic hydrocarbons and the last lower intensity peaks

consist in higher molecular weight compounds (in this case FFAs), that are strongly reduced by increasing

catalysts mass. The detected higher peaks consist in paraffinic hydrocarbons in the range of C7 - C11 and

C15, which are probably formed by the deoxygenation reaction of palmitic and oleic acid (respectively C16:0

and C18:1). In this case, the desired cracking reactions (i.e. decarboxylation) occurred in the proximity of the

carboxyl group (C17 and C15) or where the double bond is located along the chain of the fatty acid (from C7

to C10).

Fatty Acids Paraffins Olefines Aromatics Ciclo-octeneUndetected

Species

UCO AC 13% 5% 8% 8% 2% 64%

UCO AC* 7% 15% 11% 6% 3% 58%

Fatty Acids AC* 16% 20% 12% 6% 2% 44%

0%

10%

20%

30%

40%

50%

60%

70%

We

igh

t p

erc

en

tage

(w

t%)



Figure 10: GC-FID spectrum from UCO AC* test.

Comparing the range of detected hydrocarbons with traditional fossil fuels, the fraction between C8 and C12

takes interest in the production of renewable jet fuel, while the heavier and lighter fractions can be evaluated

as gasoline and diesel-range fuel. However, a large fraction of undetected oxygenated compounds were

detected into bio-oil, thus the separation of hydrocarbons from the other chemical classes represents the

next challenge.

4.3.4 Investigation of unidentified fraction of bio-oil

After the experimental campaign, RE-CORD and NESTE worked together to a more detailed

characterization of bio-oil samples. In particular, the bio-oil sample from UCO, produced at 500 °C of

pyrolysis temperature and 2.5 h-1

of WHSV, was selected as the most representative for further investigation.

NESTE received 0.5 litre sample of bio-oil from RE-CORD laboratory and focused the analysis to identify the

unidentified fraction from GC MS/FID. Here following the analysis performed by NESTE: GPC (UNI EN ISO

16931 (CEN. Animal and vegetable fats and oils - Determination of polymerized triacylglycerols by high-

performance size-exclusion chromatography HPSEC) and ESI MSMS techniques were performed to

investigate the presence of glycerides-based compounds: TriAcylGlycerides (TAG), DiAcylGlycerides (DAG),

MonoAcylGlycerides (MAG), oligomers. GPC detect also free fatty acids (FFA).

The results GPC analysis (Table 12 ) show the overall size distribution of these molecules in the sample as area%

including also compounds with similar molecular weight to those investigated, such as oxidized or oligomerized lipids.

Table 12: GPC results.

Compound Area%

TAG 4.7

DAG 16.3

MAG 8.7

FFA 69.6

OLIGOMERS (larger then TAG)

0.7

GPC analysis shows that there are only 0.72 % (area-% of the chromatogram) oligomers, which are

identified in the range of C22:0-C50:0. In addition, quite a large number of the components identified as

being the size of TAG, DAG or MAG could be actually something else: other lipid species, oxidized lipids or

oligomerized lipids. The ESI-MSMS analysis (Table 13) gives the actual amounts of MAG, DAG and TAG,

which are in total less than 0.7 wt%.



Table 13: ESI-MSMS results.

Compound wt%

TAG ≤ 0.4

DAG ≤ 0.3

MAG 0.0

In order to determine an approximate content of hydrocarbons, another set of GPC analyses was performed

after two solid phase extraction (SPE-GPC).

Two different SPE cartridges (NH2 and Si) were adopted to characterize the “neutral lipids” and FFA fraction

(including all acidic components classified as “polar lipids”). Neutral lipids was then fractionated to separate

the pure hydrocarbons. The three fractions were then analysed by GPC giving the results shown in Table 14,

which are semi-quantitative, derived from analytical standards, and affect by a relevant error of 15-20%. The

estimation of the polar components is calculated by difference.

Table 14: SPE-GPC results.

Fractions %

Hydrocarbons ≈20

Neutral fraction ≈20

Free fatty acids ≈15-20

Polar fraction by

difference ≈40

According to these data only a small amount of glycerides was detected in the sample, even if GPC suggest

the presence of glyceride sized components, probably other oxygenated lipid species, which are not

detected by ESI MSMS. The SPE-GPC identified different fractions of the sample, hydrocarbons, neutral

lipids and polar lipids and give an estimation of their distribution in the sample.

GPC or the ESI-MSMS are not able to detect hydrocarbons: they only measure the lipid-type components,

so the size distribution is only for the lipid part of the sample. GPC analysis is also used in NESTE labs to

measure the total lipid content of samples. It is not as accurate as GC-MS analysis, because length of the

fatty acids has a limited effect on the results. So the total oil content of bio-oil sample is 64 wt%, according to

the GPC measurement. The remaining part, approximately 36 wt%, is close to 35 wt% of total hydrocarbons

content, detected by GC MS-FID at RE-CORD lab, and reported in the precedent paragraph.

A final GC-simulated distillation [63], was performed to define the boiling point distribution of the total sample.

According to 76 wt% of bio-oil sample, this fraction has carbon number less than C24-C26, while 90 wt%

fraction, less than C40-C42. It seems that GPC/ESI-MSMS analysis gives too low figure for these heavier

components, because they are not anymore fatty material, but hydrocarbons.



Figure 11: GC-simulated distillation.

4.3.5 Discussion

Bio-hydrocarbons fractions were produced and detected. However, a relevant oxygenated lipid-derived

fraction was generated during catalytic cracking of pyrolysis. The best results, achieved with activated

carbons as catalyst, suggest a detailed investigation on other process parameters, such as temperatures

and WHSVs, to improve the bio-oil quality in terms of hydrocarbons and reduce the lipid-derived fraction.

Moreover, the small scale of experiments generates lower yields which can be improved by scaling up the

process.

The author studied the energy requirements for pyrolysis process of triglycerides [60]: assuming the

industrial scale of the process, it can be theoretically self-sustained adopting the 72% of the heat generated

from the combustion of the non-condensable gases.

Moreover, the evaluation of commercial process of hydrotreating could be an opportunity to upgrade the bio-

oil in order to deoxygenate the lipid-derived species, The implementation of catalytic pyrolysis into an

existing biorefinery could be considered an opportunity in terms of reduced hydrogen demand than the

current HEFA process. Moreover, once assessed the true nature of the bio-oil, some fraction could be also

evaluated for chemicals production.

0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400 500 600 700

Recovere

d m

ass [w

t%]

Boiling points [°C]



Development of UCO specifications 5

On the basis of the acquired experience during the experimental campaign of catalytic pyrolysis, a

preliminary filtration of UCO before processing does not affect the process functioning. However, the small

scale of experiments and their short duration, suggest further investigation to understand the true nature of

bio-oil and the effect of UCO on catalysts. Hydrotreating requires high quality catalysts due to presence of

noble metals [64], less tolerant to irregularly contaminated feed as UCO. In addition, the same catalysts

adopted for vegetable oil conversion, could be not effective changing some properties of feedstock,

Next paragraph investigates on the effect of catalysts deactivation and HVO properties when used cooking

oils or fatty acids are processed. The investigation of the constraints and barriers for their use will give further

detail to investigate the required quality of feedstock. Finally, a proposal of UCO specification is carried out.

5.1 Hydrotreating barriers and constraints

The role of catalysts is crucial in the production of high grade jet fuels from vegetable oils [65]. Catalysts

based on Ni and Mo supported on oxides are considered very active for the hydrotreating process whereas

Pd and Pt supported on zeolites are the most adopted hydro-isomerization catalysts. The mechanisms of

catalysts deactivation are numerous and complex [66], but their understanding permit to have fundamental

information on the quality of feedstock.

The impact of hydrotreating products originating from heteroatoms (mainly H2S, NH3 and water from S, N

and O initial content of feedstock) on the catalyst activity and selectivity was deeply studied [67]. It has been

shown that ammonia strongly inhibits both hydrodeoxygenation and (hydro)decarboxylation reactions. Water

has less effect on this side. In contrast, hydrogen sulphide promoted deoxygenation of esters over both NiMo

and CoMo catalysts [67].

Kubicka and Horàcek [68] investigated the catalytic deoxygenation of different rapeseed-oil-derived

feedstocks (i.e. used cooking oil) over sulfided CoMo catalyst, at 310 °C of process temperature and 3.5MPa

(H2 pressure), in a fixed-bed reactor, aiming at description of parameters affecting the catalyst deactivation.

Presence of alkalis promoted catalyst deactivation due to their deposition on catalyst surface leading to

blockage/poisoning of active sites. The effect of phosphorus content was two-fold: when charge-

compensating alkalis were present, corresponding phosphates were deposited above and at the beginning

of catalyst bed leading to gradual build-up of deposits. On the other hand, in absence of alkalis

decomposition of phospholipids yielded phosphoric acid that catalysed oligomerization reactions leading to

rapid catalyst deactivation by carbonaceous deposits. Presence of sulphur compounds, here demonstrated

by addition of dimethyldisulfide (DMDS), was shown to be beneficial for the catalyst deoxygenation

performance. While in absence of DMDS gradual deactivation was observed, addition of DMDS resulted in a

stable catalyst performance. The changes in active sites structure caused by their desulfurization under

hydrogen atmosphere were shown to be partially reversible. Nonetheless, part of the active sites was

irreversibly lost. In addition to modifying catalyst activity, H2S formed by DMDS decomposition, influenced

significantly the selectivity of deoxygenation reactions. The presence of H2S resulted in an augmented

(hydro)decarboxylation catalyst selectivity plausibly owing to the increased catalyst acidity.



Bezergianni et al [69] studied the effect on HDT catalysts for waste cooking oil, conducing the process in a

temperature range of 330–390 °C, under constant LHSV (1 h-1

) and a range of pressure 8.27–13.79 MPa. As

typical hydrotreating catalyst, HDT showed a drastic activity on heteroatoms removal: sulphur from 27,200

to 71.8 ppm, and nitrogen from 219.8 to 0.59 ppm (starting values were high due to presence of additives,

such as DMDS, to regulate the catalyst activity).

The effect of feedstock acidity in hydrotreating was also tested: Simakova et al [70] investigated the catalytic

deoxygenation of palmitic and stearic acids mixture over four synthesized Pd catalysts supported on

synthetic carbon (Sibunit) in a semi-batch reactor and dodecane as a solvent at 260–300 °C. The main liquid

phase products were heptadecane and pentadecane, and an optium Pd dispersion was found. The same

authors [71] tested five different fatty acids, C17–C20 and C22, as in a semibatch reactor over 1 wt% Pd/C

(Sibunit) catalyst. The same decarboxylation rates were obtained for pure fatty acids, whereas extensive

catalyst poisoning and/or sintering and coking occurred with low purity fatty acids as reactants. One reason

for catalyst poisoning using behenic acid (C22) as a feedstock was its high phosphorus content (266 mg kg-

1). The decarboxylation rate of fatty acids decreased also with increasing fatty acid to metal ratio. The

catalytic hydroconversion of C18 fatty acids was also tested for the production of iso/n-paraffins at different

temperature, WHSV, pressure and H2/oil ratio [72] and evaluating the performance of different hydrogen

donors (H-donors) [73] than pure hydrogen. Hengst et al [74] tested the gas phase deoxygenation of oleic

acid over palladium containing acidic catalysts under solvent-free condition, with a selectivity of diesel

fraction up to 90% using 1wt% Pd/Siral70 as catalyst, at 350 °C, WHSV of 1.85 s−1

and a H2 pressure of 15

bar.

So the use of hydrogen is fundamental to promote the deoxygenation of fatty acids, but selecting the most

effective catalyst at the optimum reaction conditions, a high quality product can be obtained. Na et al [75]

tested the decarboxylation of oleic acid without hydrogen using hydrotalcites, where the effect of MgO

content in hydrotalcites and reaction temperatures on the decarboxylation performance in terms of oleic acid

conversion and product distribution were investigated. However, in this case, the absence of hydrogen does

not favour the removal of other heteroatoms contained into initial feedstock. Oleic acid conversions were

more than 98% and the oxygen content in the reaction product was less than 1wt% in the case of

decarboxylation with catalysts at MgO content of 63 and 70 wt%, at 673 K.

A possible pathway to

5.2 A proposal for UCO specifications

On the basis of the feedstock requirements for catalytic pyrolysis, but more in detail, for the use of alternative

lipid-based feedstock in hydrotreating (such as UCO), a proposal of UCO specification was carried out. This

list includes limits and barriers related to filtered UCO its thermochemical processing. DIN 51623 [8] was

adopted as guideline for some parameters. The following table shows the proposal of UCO specifications.

Table 15: Esterified UCO analysis (methods of analysis are described in 4.2.3).

Parameter Unit Method ITAKA UCO

specifications

UCO filtered at 1

μm

Density kg m-3 UNI EN ISO 3675 900 - 930 911

Flashpoint °C DIN EN ISO 2719 > 220



Cetane Number - DIN EN 15195 > 39

Kinematic Viscosity

at 40 °C mm

2 s

-1 UNI EN ISO 3104 < 40 38.15

Acid value mgKOH g-1 UNI EN 14104 0 - 200 2.63

Free Fatty Acid wt% UNI EN ISO 660 0 - 100 1.31

Iodine Number g Iodine 100g-1 UNI EN 14111 < 125 93

Water content wt% UNI EN ISO 8534 < 0.2 0.080

Ash wt% UNI EN ISO 6245 < 0.01 0.01

Total contamination mg kg-1 UNI EN 12662 < 180 256

Sulphur wt% Internal method < 0.001 0

Calcium mg kg-1 UNI EN 14538 < 1 0

Magnesium mg kg-1 UNI EN 14538 < 1 0.22

Phosphorus mg kg-1 UNI EN 15290 < 3 10.1

C wt% ASTM D5291 76.3

H wt% ASTM D5291 11.7

N mg kg-1 ASTM D 4629 < 100 137

O wt% calculated

11.98

Calorific value, higher kJ kg-1 DIN 51900-2 > 36,000 38.91

Density, Iodine number, Ca and Mg content, calorific value: no significant difference were measured

in UCOs compared to vegetable oils, once it was filtered and water content removed. The UCO

specifications maintained the same limits of vegetable oils as fuel (DIN 51623).

Flash point and Cetane number: despite the UCO samples was not investigated in flash point and

Cetane number, the limits are maintained the same as DIN 51605, a less severe limits than DIN

51623.

Kinematic viscosity at 40 °C: 40 cSt was selected as limit for kinematic viscosity. The measured

value in filtered UCO (about 38 cSt) was sufficient to process the UCO in catalytic pyrolysis and

probably, in hydrotreating.

Acid value and free fatty acid: as investigated in precedent paragraph, a large number of

experiments were performed with pure fatty acids as feedstock for hydrotreating. Moreover, the

catalytic pyrolysis tests showed an improved deoxygenation than the tests with UCO. No limits of

FFAs content were assumed for UCO specifications (and the correlated acid value).

Water content: a more tolerant 0.2 than 0.075 wt% (DIN 51623) was selected as limit for moisture

content into filtered UCO. This value can be easily carried out by paper filtration.

Ash: despite ash content is not present in traditional vegetable oils, the contamination of UCO can

increase its content. A conservative limit of 0.01 wt% was adopted to preserve catalysts selectivity.

Total contaminations: during D2.3 experimental activities, an high grade filtration reduced the solid

impurities (up to 0.8 micron, according to UNI EN 12662) under 180 ppm. For this reason, a limit of

180 ppm was selected as suitable before UCO thermochemical processing.

S, P, O, N content: the removal of heteroatoms content by means of the use of hydrogen is possible

during hydrotreating. However, as reported in precedent paragraph, their presence could affect the

process. DIN 51623 limits were assumed also for UCO specifications. Despite N and P content of

filtered UCO was above these limits, the catalytic pyrolysis process was not affected by their



contaminations. Considering the hydrotreating process, depending of catalysts adopted, further

study to assess these limits are necessary.

Metals: the presence of metals content has to be reduced to avoid catalysts deactivation due to

poisoning. DIN 51623 limits were also assumed for these specifications (even for K and Na content,

above 1 ppm).



Conclusions 6

The chance to recycle a waste, such as used cooking oils, for aviation renewable fuel production represents

a promising opportunity for greening the sector. Several samples of raw UCO were collected, analysed and

evaluated in their relevant properties as feedstock for commercial hydrotreating process of vegetable oils

(D2.3). Raw UCO needs to be filtered and upgraded before its use as feedstock for thermochemical

processing. Esterification of UCO was considered as upgrading treatment: three samples of esterified UCO

were analysed and investigated as potential feedstock. However, thermochemical bio-hydrocarbons

production can be also carried out from pure fatty acids, thus esterification of UCO was not considered. High

grade filtration reduced contaminants and water content of raw UCO, restoring a large number of properties

to initial values of vegetable oils.

An experimental investigation on catalytic and non-catalytic pyrolysis of filtered used cooking oil (UCO) and

pure fatty acids (FAs) was necessary to investigate on UCO requirements as feedstock. It aimed also at

developing a preliminary screening and understanding of the process and catalysts, focusing on lipid

conversion to hydrocarbon. Representative batches of UCOs and FAs were opportunely filtered and

characterized. Four catalysts were selected and tested at 500 °C of process temperature and 4 h-1

of WHSV

in a 1.5 kg h-1

pilot catalytic pyrolysis unit. The evaluation of bio-oils properties and gas chromatographic

investigation on produced hydrocarbons indicated that, by selecting the most appropriate catalyst (activated

carbons) and process parameters, deoxygenation (decarboxylation) can be performed through catalytic

pyrolysis without hydrogen addition.

An improved experimental campaign by increasing catalyst mass (at 2.5 h-1

of WHSV) produced about 63

wt% of bio-oil (from UCO) containing the 35 wt% of hydrocarbons, and 49 wt% of bio-oil (from FAs)

containing 40 wt% of hydrocarbons. A large fraction of undetected species was found into bio-oils samples,

thus RE-CORD, in close collaboration with NESTE, selected the most suitable blend for the production of

bio-kerosene. Numerous analysis were performed on the best selected samples to identify the contained

chemical species: a mixture of lipid-derivated species.

The developed process, which was demonstrated that is theoretically self-sustained, integrates the concept

of biorefinery including the commercial process of the hydrotreating of vegetable oils (HEFA) to upgrade the

bio-oil, with an overall reduction in terms of hydrogen demand. The experience on the analysis of UCOs and

bio-oil samples, together with NESTE and literature data on hydrotreating process, permitted to identify the

critical parameters of UCO as feedstock for thermochemical processing. A proposal of UCO specification

was finally carried out.



References 7

[1] ASTM International. ASTM D4054-14, Standard Practice for Qualification and Approval of New Aviation Turbine Fuels and Fuel Additives. 2014. doi:10.1520/D4054-14.

[2] ASTM International. ASTM D7566-15c, Standard Specification for Aviation Turbine Fuel Containing Synthesized Hydrocarbons. 2015. doi:10.1520/D7566-15C.

[3] ASTM International. ASTM D1655-15de1, Standard Specification for Aviation Turbine Fuels. 2015. doi:10.1520/D1655-15DE01.

[4] Pearlson M, Wollersheim C, Hileman J. A techno-economic review of hydroprocessed renewable esters and fatty acids for jet fuel production. Biofuels, Bioprod Biorefining 2012;6:246–56. doi:10.1002/bbb.

[5] EU FP7 ITAKA. Grant Agreem No 308807 2012. http://www.itaka-project.eu.

[6] Lonza H. Hass, H. Maas, A. Reid, K.D. Rose L, Programme JECB. EC & JRC: EU renewable energy targets in 2020: Analysis of scenarios for transport fuels. 2014. doi:10.2788/74948.

[7] Buffi M, Chiaramonti D, Marsili Libelli I. D2.3 Report describing the main barriers to innovation and recommendations for further research. EU FP7 ITAKA project. 2015.

[8] DIN. Fuels for vegetable oil compatible combustion engines - Fuel from vegetable oil - Requirements and test methods. DIN 51623. Deutsches Institut Für Normung, Germany: 2015.

[9] Guerrero F. CA, Guerrero-romero A, Sierra FE. Biodiesel Production from Waste Cooking Oil. In: Stoytcheva M, Gisela M, editors. Biodiesel - Feed. Process. Technol., 2010, p. 23. doi:10.5772/25313.

[10] Knothe G, Steidley KR. A comparison of used cooking oils: a very heterogeneous feedstock for biodiesel. Bioresour Technol 2009;100:5796–801. doi:10.1016/j.biortech.2008.11.064.

[11] Canakci M. The potential of restaurant waste lipids as biodiesel feedstocks. Bioresour Technol 2007;98:183–90. doi:10.1016/j.biortech.2005.11.022.

[12] Neff W, Frankel E, Fujimoto K. Autoxidative dimerization of methyl linolenate and its monohydroperoxides, hydroperoxy epidioxides and dihydroperoxides. J Am Oil Chem … 1988;65:616–23.

[13] Directive 2008/98/EC of the European Parliament and of the Council of 19 November 2008 on waste and repealing certain Directives (Text with EEA relevance). http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32008L0098; n.d.

[14] Directive 2004/35/CE of the European Parliament and of the Council of 21 April 2004 on environmental liability with regard to the prevention and remedying of environmental damage. http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32004L0035; n.d.

[15] Regulation (EC) No 1774/2002 of the European Parliament and of the Council of 3 October 2002 laying down health rules concerning animal by-products not intended for human consumption. http://eur-lex.europa.eu/legal-content/EN/TXT/?qid=1421749928719&uri=CELEX:32002R1774; n.d.

[16] Mandolesi de Araújo CD, de Andrade CC, de Souza e Silva E, Dupas FA. Biodiesel production from used cooking oil: A review. Renew Sustain Energy Rev 2013;27:445–52. doi:10.1016/j.rser.2013.06.014.

[17] Talebian-Kiakalaieh A, Amin NAS, Mazaheri H. A review on novel processes of biodiesel production from waste cooking oil. Appl Energy 2013;104:683–710. doi:10.1016/j.apenergy.2012.11.061.

[18] Mohammadshirazi A, Akram A, Rafiee S, Bagheri Kalhor E. Energy and cost analyses of biodiesel production from waste cooking oil. Renew Sustain Energy Rev 2014;33:44–9. doi:10.1016/j.rser.2014.01.067.

[19] Leung DYC, Wu X, Leung MKH. A review on biodiesel production using catalyzed transesterification. Appl Energy 2010;87:1083–95. doi:10.1016/j.apenergy.2009.10.006.

[20] EUBIA. Transformation of Used Cooking Oil into biodiesel: From waste to resource. 2015.



[21] Toop G, Alberici S, Spoettle M, van Steen H. Trends in the UCO market. 2013.

[22] Tacconi D, Prussi M, Buffi M, Chiaramonti D. D2.7 Information related to economic , social and environmental parameters 3rd. EU FP7 ITAKA project. 2015.

[23] ISCC. Guidelines of the ISCC system are based on European directives and national (German) ordinances relevant for the production of sustainable biomass n.d. http://www.iscc-system.org/en/iscc-system/iscc-eu-and-iscc-de/legislation/.

[24] Tacconi D, Prussi M, Buffi M, Chiaramonti D. D2.6 Information related to economic , social and environmental parameters 2nd. EU FP7 ITAKA project. 2014.

[25] Bouaid A, Vázquez R, Martinez M, Aracil J. Effect of free fatty acids contents on biodiesel quality. Pilot plant studies. Fuel 2016;174:54–62. doi:10.1016/j.fuel.2016.01.018.

[26] Ragland KW, Aerts DJ, Baker a. J. Properties of wood for combustion analysis. Bioresour Technol 1991;37:161–8. doi:10.1016/0960-8524(91)90205-X.

[27] Buffi M, Rizzo AM, Prussi M, Chiaramonti D. Catalytic pyrolysis of used cooking oils: a novel route to alternative jet fuel. Eur. Biomass Conf. Exhib., Wien: ETA Florence; 2015, p. 27.

[28] Sotelo-Boyás R, Trejo-Zárraga F, Hernández-Loyo F de J. Hydroconversion of Triglycerides into Green Liquid Fuels. InTech, 2012, p. 30. doi:http://dx.doi.org/10.5772/48710.

[29] Buffi M, Rizzo AM, Prussi M, Chiaramonti D. Strategies to meet the quality specification of biofuels for lipid-based feedstock. Euro Fed Lipid, Florence: 2015, p. 27.

[30] Satyarthi JK, Chiranjeevi T, Gokak DT, Viswanathan PS. An overview of catalytic conversion of vegetable oils/fats into middle distillates. Catal Sci Technol 2013;3:70–80. doi:10.1039/C2CY20415K.

[31] Huber GW, O’Connor P, Corma A. Processing biomass in conventional oil refineries: Production of high quality diesel by hydrotreating vegetable oils in heavy vacuum oil mixtures. Appl Catal A Gen 2007;329:120–9. doi:10.1016/j.apcata.2007.07.002.

[32] Chiaramonti D, Prussi M, Buffi M, Tacconi D. Sustainable bio kerosene: Process routes and industrial demonstration activities in aviation biofuels. Appl Energy 2014;136:767–74. doi:10.1016/j.apenergy.2014.08.065.

[33] Jain S, Sharma MP. Biodiesel production from Jatropha curcas oil. Renew Sustain Energy Rev 2010;14:3140–7. doi:10.1016/j.rser.2010.07.047.

[34] Johnson MC. Hydrothermal processing of high-lipid biomass to fuels. Massachusetts Institute of Technology, 2012.

[35] Maher KD, Bressler DC. Pyrolysis of triglyceride materials for the production of renewable fuels and chemicals. Bioresour Technol 2007;98:2351–68. doi:10.1016/j.biortech.2006.10.025.

[36] Chang CC, Wan SW. China’s motor fuels from tung oil. Ind Eng Chem 1947;39:1543–8. doi:10.1021/ie50456a011.

[37] Alencar JW, Alves PB, Craveiro a a. Pyrolysis of Tropical Vegetable Oils. J Agric Food Chem 1983;31:1268–70. doi:10.1021/jf00120a031.

[38] Schwab a W, Dykstrab GJ, Selkeo E, Sorensonb SC, Prydeo EH. Diesel Fuel from Thermal Decomposition of Soybean Oil. J Am Oil Chem Soc 1988;65:1781–6. doi:10.1007/BF02542382.

[39] Idem RO, Katikaneni SPR, Bakhshi NN. Catalytic conversion of canola oil to fuels and chemicals: roles of catalyst acidity, basicity and shape selectivity on product distribution. Fuel Process Technol 1997;51:101–25. doi:10.1016/S0378-3820(96)01085-5.

[40] Tani H, Hasegawa T, Shimouchi M, Asami K, Fujimoto K. Selective catalytic decarboxy-cracking of triglyceride to middle-distillate hydrocarbon. Catal Today 2011;164:410–4. doi:10.1016/j.cattod.2010.10.059.

[41] Teixeira V, Sousa LA. The Role of Catalysis for the Sustainable Production of Bio-fuels and Bio-chemicals. Role Catal. Sustain. Prod. Bio-fuels Bio-chemicals, vol. 1, Elsevier; 2013, p. 67–92. doi:10.1016/B978-0-444-56330-9.00003-6.



[42] Zhenyi C, Xing J, Shuyuan L, Li L. Thermodynamics Calculation of the Pyrolysis of Vegetable Oils. Energy Sources 2004;26:849–56. doi:10.1080/00908310490465902.

[43] Zhang Y, Wang X, Li Q, Yang R, Li C. A ReaxFF Molecular Dynamics Study of the Pyrolysis Mechanism of Oleic-type Triglycerides. Energy & Fuels 2015:150729145212006. doi:10.1021/acs.energyfuels.5b00720.

[44] ASTM. Standard Test Methods for Instrumental Determination of Carbon, Hydrogen, and Nitrogen in Petroleum Products and Lubricants D5291 – 10. ASTM. West Conshohocken, PA (USA): American Society for Testing and Materials: 2011.

[45] ASTM. Standard Test Method for Trace Nitrogen in Liquid Petroleum Hydrocarbons by Syringe/Inlet Oxidative Combustion and Chemiluminescence Detection ASTM D4629. ASTM, West Conshohocken, PA (USA): American Society for Testing and Materials: 2011.

[46] DIN. Determining the gross calorific value of solid and liquid fuels using the isoperibol or static-jacket calorimeter, and calculation of net calorific value. DIN 51900-2. DIN. Beuth Verlag GmbH, 10772 Berlin, Germany, Deutsches Institut Für Normung e.V.: 2003.

[47] CEN. Determining the gross calorific value of solid and liquid fuels using the bomb calorimeter, and calculation of net calorific value. DIN 51900-1. DIN. Beuth Verlag GmbH, 10772 Berlin, Germany, Deutsches Institut Für Normung e.V.: 2000.

[48] CEN. Determining Animal and vegetable fats and oils. Determination of acid value and acidity. EN ISO 660. Brussels, Belgium: European Committee for Standardization: 2009.

[49] CEN. Fat and oil derivatives - Fatty Acid Methyl Esters (FAME) - Determination of acid value EN 14104. Brussels, Belgium: European Committee for Standardization: 2003.

[50] CEN. Determination of kinematic viscosity and calculation of dynamic viscosity of transparent and opaque liquids DIN EN ISO 3104. Brussels, Belgium: European Committee for Standardization: 1996.

[51] CEN. Animal and vegetable fats and oils. Determination of water content Karl Fischer method (pyridine free) EN ISO 8534. Brussels, Belgium: European Committee for Standardization: 2008.

[52] Oasmaa A, Peacocke C. Properties and fuel use of biomass-derived fast pyrolysis liquids. A guide. VTT Publ 2010;731:79 p. + app. 46 p.

[53] CEN. Animal Fat and oil derivates Fatty acid methyl ester (FAME) Determination of Ca, K, Mg and Na content by optical emission spectral analysis with inductively coupled plasma (ICP OES) EN 14538. Brussels, Belgium: European Committee for Standardization: 2006.

[54] CEN. Solid biofuels. Determination of major elements Al, Ca, Fe, Mg, P, K, Si, Na and Ti EN 15290. Brussels, Belgium: European Committee for Standardization: 2011.

[55] CEN. Liquid petroleum products –Determination of contamination in middle distillates DIN EN 12662. Brussels, Belgium: European Committee for Standardization: 2008.

[56] CEN. Crude petroleum and liquid petroleum products. Laboratory determination of density. Hydrometer method EN ISO 3675. . Brussels, Belgium: European Committee for Standardization: 2008.

[57] CEN Animal and vegetable fats and oils - Determination of insoluble impurities content. EN ISO 663. Brussels, Belgium: European Committee for Standardization: 2008.

[58] CEN. Animal Fat and oil derivatives Fatty Acid Methyl Esters (FAME) Determination of iodine value. EN 14111. Brussels, Belgium: European Committee for Standardization: 2003.

[59] CEN. Animal and vegetable fats and oils.Preparation of methyl esters of fatty acids EN ISO 5509. ISO. CaBrussels, Belgium: European Committee for Standardization: 2000.

[60] Buffi M. Catalytic pyrolysis of used cooking oils: a novel route to renewable hydrocarbons for transport fuel production. University of Florence, 2016. doi:(under publication).

[61] Chiaramonti D, Buffi M, Rizzo AM, Prussi M, Martelli F. Bio-Hydrocarbons through Catalytic Pyrolysis of Used Cooking Oils: Towards Sustainable Jet and Road Fuels. Energy Procedia 2015;82:343–9. doi:10.1016/j.egypro.2015.11.800.



[62] Chiaramonti D, Buffi M, Rizzo AM, Lotti G, Prussi M. Bio-Hydrocarbons through Catalytic Pyrolysis of used cooking oils and fatty acids for sustainable jet and road fuel production (Under publication). Biomass and Bioenergy 2016.

[63] Vickers AK. Higher-Temperature Simulated Distillation with DB-HT Sim Dis Columns Application. Agilent Technologies, Inc. 91 Blue Ravine Road Folsom, CA 95630: 2002.

[64] Kordulis C, Bourikas K, Gousi M, Kordouli E, Lycourghiotis A. Development of nickel based catalysts for the transformation of natural triglycerides and related compounds into green diesel: A critical review. vol. 181. Elsevier B.V.; 2016. doi:10.1016/j.apcatb.2015.07.042.

[65] Galadima A, Muraza O. Catalytic upgrading of vegetable oils into jet fuels range hydrocarbons using heterogeneous catalysts: A review. J Ind Eng Chem 2015;29:12–23. doi:10.1016/j.jiec.2015.03.030.

[66] Bartholomew CH. Mechanisms of catalyst deactivation. Appl Catal A Gen 2001;212:17–60. doi:10.1016/S0926-860X(00)00843-7.

[67] Laurent E, Delmon B. Study of the hydrodeoxygenation of carbonyl, carboylic and guaiacyl groups over sulfided CoMo/γ-Al2O3 and NiMo/γ-Al2O3 catalyst. Appl Catal A Gen 1994;109:97–115. doi:10.1016/0926-860X(94)85005-4.

[68] Kubicka D, Horácek J. Deactivation of HDS catalysts in deoxygenation of vegetable oils. Appl Catal A Gen 2011;394:9–17. doi:10.1016/j.apcata.2010.10.034.

[69] Bezergianni S, Kalogianni A, Dimitriadis A. Catalyst evaluation for waste cooking oil hydroprocessing. Fuel 2012;93:638–41. doi:10.1016/j.fuel.2011.08.053.

[70] Simakova I, Simakova O, Mäki-Arvela P, Simakov A, Estrada M, Murzin DY. Deoxygenation of palmitic and stearic acid over supported Pd catalysts: Effect of metal dispersion. Appl Catal A Gen 2009;355:100–8. doi:10.1016/j.apcata.2008.12.001.

[71] Simakova I, Simakova O, Maki-Arvela P, Murzin DY. Decarboxylation of fatty acids over Pd supported on mesoporous carbon. Catal Today 2010;150:28–31. doi:10.1016/j.cattod.2009.07.064.

[72] Yang Y, Wang Q, Zhang X, Wang L, Li G. Hydrotreating of C18 fatty acids to hydrocarbons on sulphided NiW/SiO2-Al2O3. Fuel Process Technol 2013;116:165–74. doi:10.1016/j.fuproc.2013.05.008.

[73] Yang Y, Wang Q, Chen H, Zhang X. Enhancing selective hydroconversion of C18 fatty acids into hydrocarbons by hydrogen-donors. Fuel 2014;133:241–4. doi:10.1016/j.fuel.2014.05.044.

[74] Hengst K, Arend M, Pf??tzenreuter R, Hoelderich WF. Deoxygenation and cracking of free fatty acids over acidic catalysts by single step conversion for the production of diesel fuel and fuel blends. Appl Catal B Environ 2015;174-175:383–94. doi:10.1016/j.apcatb.2015.03.009.

[75] Na J-G, Yi BE, Kim JN, Yi KB, Park S-Y, Park J-H, et al. Hydrocarbon production from decarboxylation of fatty acid without hydrogen. Catal Today 2010;156:44–8. doi:10.1016/j.cattod.2009.11.008.

[76] DIN. Fuels for vegetable oil compatible combustion engines - Fuel from rapeseed oil - Requirements and test methods- DIN 51605. Deutsches Institut Für Normung, Germany: 2010.

Documents

D2.8 Development of UCO specificationsvegetable oil is suitable for commercial process of hydrotreating with no precautions than other traditional vegetable oils. Used cooking oil